Executive Management board member

Leading AI team at food recognition project. Both team management (Agile, including hiring) and research/development responsibilities. Main achievements: - Successfully built the whole model preparation pipeline including collecting videos, processing them in order to extract the most useful frames, data filtering, model training, model testing, and model deployment. - Prepared data annotation protocol. - Prepared manual data capturing protocol for more efficient data collection. - Moved the whole processing pipeline from local servers to AWS. - Did a very smooth pilot at Ikea Italy with >500 customers per day for 6 months which proved all the positive business effects.

Project analysis and estimation. Development and implementation of machine learning/data analysis/computer vision algorithms. Projects (all successfully finished): - tennis match analysis for active game fragments extraction via player and ball behavior analysis; - photo-based pig weight prediction; - multiple athletes tracking; - news recommender system; - elite horse classification.

Building the AI team from scratch. Development and implementation of machine learning/computer vision algorithms. Project and product management, data scientists and data engineers hiring, agile team management. Main achievements: - Developed an algorithm for video and photo analysis for tag and genre assignment using deep neural nets with >90% accuracy; - Developed and deployed a real-time video tagging via face recognition with a flexible person database (about 10k with >98% accuracy at the time I've left); - Built and deployed a recommendation system based on user patterns and tracks which resulted in 78% traffic increase in the recommendation tab; - Developed and deployed spam and abuse detection system which was able to detect >1000 spam accounts per day with <0.1% of false positives.

Conducting a research work in computer vision and signal processing. Writing papers, applying for grants, supervising master and bachelor thesis, lecturing. Projects: multimodal image registration, radar signal processing, object detection and tracking, computer vision-based navigation.

Participating in the research project of the lab dealing with EMG control. This work focuses on the increase of the controllability of a wearable prosthetic device providing it with a more natural human-like behavior on grasping motions. My role in the project consisted in applying computer vision techniques to video data acquired by the camera mounted in glasses in order to increase the control quality.

Development of hardware and software in the fields of computer vision and signal processing. Prototyping algorithms, give instructions and tasks, consulting. Projects: - biometric suitcase lock (for the industry), - fundus image analysis (for the industry), - human face analysis; - bank clients’ info analysis (for the industry); - X-ray image analysis, - satellite image registration, - radar signal analysis.

Выполнение грантов и хоздоговорных НИР, проведение занятий

Радиотехнический, Математическое моделирование, численные методы и комплексы программ

Радиотехнический факультет, Радиотехника